This proposal represents a systematic investigation of the effects of repetition on learning and memory in amnesic patients. Although it is clear that repetition provides one of the most powerful methods to enhance memory in amnesia, little is known about the kinds of repetition that are most beneficial, and the mechanisms by which different forms of repetition have their effect. The present studies are guided by a framework that distinguishes between a fast learning system mediated by the medial temporal lobes and a slow learning system mediated by neocortex, and are based on the assumption that repetition has distinct effects in each of these systems. The first section of the proposal examines repetition in the fast medial temporal learning system by evaluating the operation in amnesic patients of three different mechanisms thought to contribute to repetition effects in individuals with intact memory function, namely (1) encoding; (2) time in working memory; and (3) study-phase retrieval. We predict that these mechanisms will be impaired in individuals with amnesia, but that conditions can be created that minimize these impairments, thus optimizing their performance. The second section of the proposal concerns the effects of repetition in the slow neocortical learning system. Little is known about the nature and characteristics of repetition in the slow learning system, because previous studies have failed to isolate the contribution of slow neocortical learning from that of fast medial temporal learning. Here we make use of a novel paradigm that allows assessment of the slow neocortical learning system in isolation and test the hypothesis that such learning is intact in amnesic patients. We also examine how various encoding and retrieval conditions affect repetition-based slow learning, with the aim of uncovering how such learning can be optimized in amnesic patients. A better understanding of the ways in which repetition operates within the fast and the slow learning systems will elucidate the operation of these memory systems and will provide a theoretical basis for the development of specific strategies aimed at enhancing distinct forms of memory in amnesia. [unreadable] [unreadable] [unreadable]